Preliminary research conducted by our group has found a powerful association between female breast cancer and body burden of DDE, the principal metabolite of DDT. This association is stronger (RR equals approximately 4) than that between breast cancer and obesity (RR less than 1.5) or delayed child-bearing (RR less than 2). Lower, but elevated, risks were found with PCBs. These exposures are preventable. To extend our earlier findings and to elucidate underlying biological mechanisms, a case-control study is proposed. We will investigate breast cancer risk among women in the New York metropolitan area, including minorities, with respect to environmental factors, carcinogen metabolizing enzymes, and oncogenes. The study is large enough to examine interactions of these factors and to examine geographic and ethnic contributions. Levels of organochlorines (DDE and PCB) in blood serum will be compared between breast cancer cases and two control groups. The relationship of breast cancer to genetic polymorphisms in carcinogen metabolizing genes (CYP1A1, CYP1A2, CYP2E1, GST) will be determined. Interactions of genotype with environmental exposures and carcinogen metabolizing genes will be investigated with respect to risk of breast cancer. Expression of p53, c- myc and erbB-2 in relation to organochlorine exposure and carcinogen- metabolizing genotype will be assessed. Estrogen receptor and epidermal growth factor receptor levels will be determined in an effort to determine relationships between breast cancer risk and environmental exposures, cytochrome P450 metabolism and oncogene expression. In order to verify the validity of blood serum levels as an indicator of body burden, organochlorine levels in mammary or upper abdominal adipose will be compared with levels in blood serum for a subset of each case and control group. A case-control study has been designed to compare 210 patient cases of invasive cancer, including 140 (66%) minority women, with 2 control groups. The control groups comprise 210 patients that required biopsy for non-hyperplastic benign breast disease and 210 women hospitalized for elective abdominal surgery (noncancer). Cases and controls will be frequency-matched on age and race. Residence status in New York City, Long Island, and New Jersey will be a variable in the analysis. Patients will be drawn from over 1000 breast biopsies, including more than 300 new breast cancer patients, seen annually in the Mount Sinai Breast Service. In addition to the effect variables needed to assess the hypotheses, data will include potential confounders for both breast cancer and organochlorine environmental exposures. Statistical analyses will be designed to model cancer risk associated with environmental exposures and their interactions with genetic and biologic markers.